Basic Types

S++ has a number of basic types, which are the building blocks of all S++ programs. This section will detail the basic types in S++. The basic types in S++ are:

• Number types

• Boolean type

• Void type

As every type in S++ is treated as first class, these types are also treated as objects, and follow standard type identifier regex, rather that a keyword-like identifier. All of these types have std::copy::Copy superimposed over them, to allow easy use of the numbers without having to place .clone() everywhere for simple calculations. Furthermore, simple CPU instructions are used to copy these types, which fit directly inside the registers. The superimposition of the Copy type is used to model tje copy logic for the memory checker.

Number Types

S++ uses a large range of numeric types, allowing for low-level optimizations and refined resource utilization on embedded devices. The naming convention is inspired by Rust, but modified to match the syntactic type-identifier requirements.

All of these types are found in the std::number module of the STL:

Type	Postfix	Size (bits)	Min	Max	Precision	Description
F8	_f8	8	~ -2^7	~ 2^7	e4m3	Signed 8-bit float
F16	_f16	16	~ -2^15	~ 2^15	e5m10	Signed 16-bit float
F32	_f32	32	~ -2^127	~ 2^127	e8m23	Signed 32-bit float
F64	_f64	64	~ -2^1023	~ 2^1023	e11m52	Signed 64-bit float
F128	_f128	128	~ -2^16383	~ 2^16383	e15m112	Signed 128-bit float
S8	_s8	8	-2^7	2^7 - 1	7 bits	Signed 8-bit integer
S16	_s16	16	-2^15	2^15 - 1	15 bits	Signed 16-bit integer
S32	_s32	32	-2^31	2^31 - 1	31 bits	Signed 32-bit integer
S64	_s64	64	-2^63	2^63 - 1	63 bits	Signed 64-bit integer
S128	_s128	128	-2^127	2^127-1	127 bits	Signed 128-bit integer
S256	_s256	256	-2^255	2^255-1	255 bits	Signed 256-bit integer
SSize	_sz	Here	Here	Here	Here	Signed “size” type
U8	_u8	8	0	2^8 - 1	8 bits	Unsigned 8-bit integer
U16	_u16	16	0	2^16 - 1	16 bits	Unsigned 16-bit integer
U32	_u32	32	0	2^32 - 1	32 bits	Unsigned 32-bit integer
U64	_u64	64	0	2^64 - 1	64 bits	Unsigned 64-bit integer
U128	_u128	128	0	2^128 - 1	128 bits	Unsigned 128-bit integer
U256	_u256	256	0	2^256 - 1	256 bits	Unsigned 256-bit integer
USize	_uz	Here	Here	Here	Here	Unsigned “size” type
BigInt	N/A	inf	-inf	inf	inf	Arbitrary precision integer
BigDec	N/A	inf	-inf	inf	inf	Arbitrary precision float

A number literal’s is always the std::bignum::bigint::BigInt or std::bignum::bigdec::BigDec if no postfix type is provided. Use the above postfixes to set the numeric type explicitly. All the number types, except BigInt and BigDec have compiler builtin operation methods, to hook into llvm-specialized functions.

The Copy class is superimposed over all of these number types so that they can be used in calculations easily without having to manually clone them. Also, numer types are small so this doesnt create a memory issue.

USize, SSize

The USize and SSize types are slightly special, as they are used to represent the numeric type that corresponds to the maximum size of addressable memory on the target architecture, unsigned and signed. This means that on a 32-bit architecture, USize <=> U32 and SSize <=> S32, and on a 64-bit architecture, the 64-bit integer equivalents. The USize type is used for indexing and slicing operations, typically seen in arrays and vectors. It is also used for system resource objects, such as socket and thread identifiers.

Boolean Type

The boolean type is S++ is std::boolean::Bool. This is a compiler known type that maps to the llvm bool type. Literals trueand false are used to create boolean value types. The Copy class is superimposed over the boolean type as-well.

Void Type

The void type in S++ is std:void::Void. This is a compiler known type that maps to the llvm void type. The Void type is used to represent the absence of a value, and is used as the return type for functions that do not return a value. Variables cannot hold a Void type, and function parameters whose type is generic (and the generic argument is Void), are removed from the substituted signature.